Language: 日本語English

zyte-web-data Plugin

Author
Zyte
Category
Automation
Topics
Web Search, Research & Learning · Claude Code Customization & Workflow · AI Agents & AI App Development
First seen
2026-07-09
Last confirmed
2026-07-13
Explanation last updated
2026-07-12

The explanation below is AI-generated. Please verify it against the sources.

zyte-web-data is a Claude Code plugin, published by Zyte, that turns a plain-English scraping prompt into a working Scrapy spider. According to the README, you give it a URL and describe what to extract, and it handles site exploration, schema discovery, code generation with web-poet page objects, and smoke testing. It proposes an extraction schema for your approval before any code is generated, and it works on sites with repeating structured content such as product catalogs, job listings, and recipes. The resulting project is a standard Scrapy project that runs locally via uv, and it can optionally be deployed to Zyte's Scrapy Cloud for scheduled runs and monitoring. A Zyte account is required only for Scrapy Cloud deployment or for using Zyte API to access sites that block standard scrapers.

Overview

Zyte provides Zyte API, used for accessing sites that block standard scrapers and for headless browser rendering, and Scrapy Cloud, a hosted service for deploying, scheduling, and monitoring Scrapy spiders that includes a free tier per the README. The plugin can optionally connect the generated spider to these services, but neither is required to run the spider locally.

What you can do with zyte-web-data

  • Generate a Scrapy spider with web-poet page objects from a plain-English prompt and a target URL
  • Review and approve a discovered extraction schema before any code is generated
  • Have the five-stage pipeline (scrape-define, scrape-spec, scrape-ensure-project, scrape-codegen, scrape-create-spider) run automatically
  • Verify extraction with an automatic smoke test and generated pytest fixtures
  • Deploy the project to Scrapy Cloud, schedule spiders, and view jobs, items, or logs via /scrape-scrapy-cloud
  • Set up Zyte account credentials via /scrape-zyte-login
  • Run the generated spider locally without Claude Code using uv run scrapy crawl

Sources

Original description (English)

Web scraping skills for Claude Code powered by the Zyte API — scrape sites, generate and run Scrapy spiders, define extraction schemas, and ship to Scrapy Cloud.

History of zyte-web-data

Back to list